[go: up one dir, main page]

WO1999038299A1 - Techniques de conversion de modules multiples pour modem mic - Google Patents

Techniques de conversion de modules multiples pour modem mic Download PDF

Info

Publication number
WO1999038299A1
WO1999038299A1 PCT/US1999/001559 US9901559W WO9938299A1 WO 1999038299 A1 WO1999038299 A1 WO 1999038299A1 US 9901559 W US9901559 W US 9901559W WO 9938299 A1 WO9938299 A1 WO 9938299A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal point
mmc
signal
transmission
values
Prior art date
Application number
PCT/US1999/001559
Other languages
English (en)
Inventor
Zhenyu Zhou
Xuming Zhang
Yuanjie Chen
Original Assignee
Conexant Systems, Inc.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Conexant Systems, Inc. filed Critical Conexant Systems, Inc.
Publication of WO1999038299A1 publication Critical patent/WO1999038299A1/fr

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/38—Synchronous or start-stop systems, e.g. for Baudot code
    • H04L25/40—Transmitting circuits; Receiving circuits
    • H04L25/49—Transmitting circuits; Receiving circuits using code conversion at the transmitter; using predistortion; using insertion of idle bits for obtaining a desired frequency spectrum; using three or more amplitude levels ; Baseband coding techniques specific to data transmission systems
    • H04L25/4917—Transmitting circuits; Receiving circuits using code conversion at the transmitter; using predistortion; using insertion of idle bits for obtaining a desired frequency spectrum; using three or more amplitude levels ; Baseband coding techniques specific to data transmission systems using multilevel codes
    • H04L25/4927—Transmitting circuits; Receiving circuits using code conversion at the transmitter; using predistortion; using insertion of idle bits for obtaining a desired frequency spectrum; using three or more amplitude levels ; Baseband coding techniques specific to data transmission systems using multilevel codes using levels matched to the quantisation levels of the channel

Definitions

  • the present invention relates generally to the use of multiple modulus conversion (MMC) techniques for mapping data in a pulse code modulation (PCM) modem system.
  • MMC multiple modulus conversion
  • PCM pulse code modulation
  • the present invention relates to techniques that contemplate transmit power parameters where MMC mapping procedures are employed.
  • FIG. 1 depicts a conceptual diagram of a typical 56 kbps communication path using current PCM modem technology.
  • a central site such as an internet service provider (ISP) 1 00, is digitally connected to a telephone network 1 30 through a transmitter 1 1 0 and a receiver 1 20 of an ISP modem 105.
  • the network 1 30 is connected to a local loop 1 50 through a central office line card 140.
  • the line card typically has a PCM codec implemented therein.
  • the local loop 1 50 is connected to the user's personal computer (PC) 1 70 at the user's site through the user's modem 1 60.
  • PC personal computer
  • the connection between the ISP modem transmitter 1 10 to the telephone network 1 30 is a digital connection with a typical data rate of about 64 kbps. Since the parameters of the telephone network 1 30 and line card 140 are dictated and set by the operating specifications of the network (and particularly the use of the ⁇ -law or A-law signal point constellations), the central site transmitter 1 1 0 is configured to transmit the digital data in a particular way to fully exploit its digital connection to the network.
  • a proposed operating protocol for 56 kbps PCM modem systems calls for the use of a data transmission scheme that performs multiple modulus conversion (MMC) on a number of bits that may be provided by a suitable scrambler or other processing element.
  • MMC multiple modulus conversion
  • Betts et al., U.S. Pat. No. 5,475,71 1 , issued December 1 2, 1 994 and Betts et al., U.S. Pat. No. 5,684,834, issued November 4, 1 997 disclose data transmission systems that utilize single modulus converters in different contexts.
  • MMC is a known technique for expressing an integer number as a sum of quotients containing multiple moduli (or bases).
  • the multiple moduli M (which are determined prior to mapping) are used to map a number of bits expressed as an integer number R.
  • MMC operates in the absence of sign bits; sign bits are removed prior to the MMC process and replaced afterward.
  • the MMC process generates the values of K ; , which represent mapping indices associated with the respective signal point constellations.
  • K represents mapping indices associated with the respective signal point constellations.
  • each K, value is used to select a particular PCM codeword for transmission over the telephone network.
  • the original digital data is recovered (assuming no transmission errors).
  • One current 56 kbps modem system utilizes signal point constellations with codewords having descending magnitudes relative to the index values.
  • the magnitudes of the codewords associated with constellation M- decrease as the value of index K-, increases.
  • the K j values are derived by the MMC procedure in a manner such that, for any given constellation, the lower values for a given index K ; may occur more frequently than higher values for that index. Consequently, for a modem system that uses a descending magnitude scheme, codewords requiring relatively higher transmit power are transmitted more often than codewords requiring relatively lower transmit power.
  • the high transmission probability associated with the larger magnitude codewords results in a communication system having an inefficient allocation of allowable transmit power.
  • the total average transmit power may be calculated to ensure that it does not exceed the power limit.
  • the total average transmit power associated with a 56 kbps modem system may not be readily obtainable.
  • the total average transmit power is often conservatively estimated for purposes of comparison to the power limit.
  • conservative estimates may result in the use of less than optimal constellations, lower system SNR, smaller minimum distances between constellation points, and a higher likelihood of errors.
  • the symbol associated with the M 0 -point constellation is transmitted first in time, the symbol associated with the
  • M 5 -point constellation is transmitted last in time, and the other symbols follow in a like temporal order.
  • this arrangement is logical and easy to implement, it may not be suitable in situations where specific cost functions, such as transmit power, are to be considered.
  • specific cost functions such as transmit power
  • following the proposed rigid transmit order may not adequately take advantage of, or compensate for, the presence of robbed bit signaling (RBS) within the telephone network. Consequently, it would be desirable to implement improved techniques in conjunction with MMC processes to address the above shortcomings of proposed 56 kbps modem systems.
  • An advantage of the present invention is that it provides improved techniques for use in a modem system that employs a multiple modulus conversion (MMC) procedure.
  • MMC multiple modulus conversion
  • signal point constellations are arranged to receive the output indices generated by the MMC procedure such that the probability of transmission of higher power codewords is relatively low compared to the probability of transmission of lower power codewords, while maintaining the same or larger minimum distance between signal points.
  • a further advantage of the present invention is that it provides a technique for obtaining the probability of occurrence for each constellation or signal point associated with a given data communication session.
  • the present invention provides a method for determining the theoretical average transmit power associated with a given set of signal point constellations and MMC parameters.
  • FIG. 1 is a conceptual block diagram of an exemplary data communication system using PCM techniques
  • FIG. 2 is a conceptual block diagram of an exemplary encoder that may be used in the ISP modem shown in FIG. 1 ;
  • FIG. 3 is a flow diagram of an exact power determination process that may be performed by the data communication system.
  • FIG. 4 is a flow diagram of an estimated power determination process that may be performed by the data communication system.
  • the present invention may be described herein in terms of functional block components and various processing steps. It should be appreciated that such functional blocks may be realized by any number of hardware components configured to perform the specified functions. For example, the present invention may employ various integrated circuit components, e.g., memory elements, digital signal processing elements, look-up tables, and the like, which may carry out a variety of functions under the control of one or more microprocessors or other control devices.
  • the present invention may be practiced in any number of data communication contexts and that the modem system described herein is merely one exemplary application for the invention. Further, it should be noted that the present invention may employ any number of conventional techniques for data transmission, control signaling, signal processing and conditioning, and the like. Such general techniques are known to those skilled in the art and will not be described in detail herein.
  • an exemplary 56 kbps PCM-based modem transmitter includes an encoder 200, which is generally configured to convert a number of input bits (identified as bits b 0 - b B ..,) into PCM codewords.
  • the PCM codewords (identified as PCM 0 - PCM 5 ) are transmitted in a serial manner through telephone network 1 30 as 8-bit octets or symbols.
  • each data frame is represented by six symbols, however, the present invention is not limited to any specific frame length, symbol size, or number of bits processed by encoder 200.
  • Encoder 200 preferably includes a modulus encoder 204 configured to perform an MMC procedure on the digital data input thereto.
  • a modulus encoder 204 configured to perform an MMC procedure on the digital data input thereto.
  • B is the number of bits that are input into modulus encoder 204 for a given data frame. As shown in FIG. 2, a number of bits (S 0 -S s . ⁇ ) are removed from the input of scrambled data bits such that the MMC procedure is only performed on information bits used to map the signal point magnitudes. Following the MMC procedure, sign bits may be assigned to each symbol in a suitable manner.
  • several operating parameters may be predetermined by the modem system prior to operation in a normal data mode. For example, the number of bits input into modulus encoder 204 may depend upon a current data rate and the number of redundant sign bits used during data transmission. Both of these parameters may be adaptively established between ISP modem 105 and user modem 1 60 during an initialization period. In one proposed 56 kbps operating scheme, B (the number of incoming bits) may range between 1 5 and
  • B may vary depending on the specific application and the present invention is not limited to any particular range of B.
  • the moduli values (M 0 - M 5 ) may be determined, along with their corresponding signal point constellations, during an adaptive initialization period. A number of factors may govern the selection of the M,
  • the exemplary MMC process described herein requires that 2 B ⁇ M Q M T . . . M 5 , which ensures that the integer represented by the B bits can be expressed in terms of the moduli.
  • the signal points selected for each individual constellation may be limited to a subset of standard values, e.g., ⁇ -law values in the United States and A-law values in Europe.
  • encoder 200 may begin transmitting information in a normal data mode. It should be noted that each of the above parameters may remain unchanged for the duration of any particular communication session and that one or more of such parameters may be modified in response to, e.g., renegotiation procedures, loss of synchronization, or the like.
  • encoder 200 includes a moduli reordering element 220, which may be configured to perform any number of optimized reordering functions with respect to the M ; values.
  • '-M 5 ', may reflect any suitable reordering of M, -l ⁇ ⁇
  • the current inventors have recognized that the order associated with the various M, values need not be rigidly designated or restricted in any manner. Indeed, this flexibility enables moduli reordering element 220 to optimally designate certain signal point constellations for use with transmissions of specific symbol locations within the given data frame.
  • moduli reordering element 220 may be dictated by any number of cost functions associated with the modem system. For any given communication session, moduli reordering element 220 may be adaptively adjusted during an initialization or renegotiation period to facilitate an optimized MMC procedure relative to one or more operating parameters. It should be appreciated that moduli reordering element 220 may be realized by any number of suitable functional components, e.g., look-up tables, programmable memory elements, or digital processing components.
  • an adaptive constellation design procedure may be able to generate constellations having larger minimum distances between signal points without exceeding the transmit power limits; the larger minimum distances are desirable to increase the transmission SNR.
  • An exemplary application of this reordering may be applied to compensate for the presence of RBS.
  • RBS occurs within a given network link, where the least significant bit of every sixth symbol is robbed and utilized for network signaling. In other words, the robbed bits are not available for use with the modem system.
  • the use of smaller M ; maps for the mapping of RBS phase symbols may be called for by an adaptive constellation procedure.
  • the inventors have discovered that power reduction can be achieved by mapping the RBS phases at the end of the MMC operation, regardless of which phase the RBS occurs.
  • the moduli associated with the RBS phases in a conventional MMC arrangement are reordered such that they are applied at the later stages (for an example where only one RBS phase is present, the reordering causes the MMC procedure to map the RBS phase at the last MMC stage) of the MMC operation.
  • the specific ordering described herein is merely exemplary and the actual reordering algorithm may be configured to optimize any parameter for the particular modem system.
  • RBS may affect more than one symbol within a data frame because different network links may utilize RBS.
  • RBS may occur in zero, one, two, or three symbols within a data frame.
  • the occurrence of RBS remains periodic regardless of the number of symbols tainted by RBS. Therefore, in accordance with the principles of the current invention, moduli reordering element 220 may be configured to suitably reorder more than one moduli to compensate for multiple RBS
  • moduli reordering element 220 may perform the following function to ensure that the smallest constellations are utilized for encoding and decoding at the RBS phases:
  • reordering element 220 may assign M 4 to M 3 ' and M 5 to M, ' .
  • moduli reordering element 220 may be governed by any suitable criteria. It should be appreciated that moduli reordering element 220 need not reorder each and every moduli (as demonstrated in the above example where M 0 remains at the initial symbol position, i.e., M i$ "reordered" to the M 0 ' position.
  • the present invention may employ adaptive optimization routines to specifically dictate a reordering that satisfies one or more cost function considerations such as the presence of RBS and transmit power limits.
  • User modem 1 60 may perform the appropriate analyses, determine a suitable reordering scheme, and transmit data indicative of the reordering scheme to ISP modem 105.
  • Encoder 200 receives the B bits to be sent during a data frame of, e.g., six symbols.
  • the MMC procedure allows the B bits to be mapped to six indices, equivalent to B/6 bits per symbol (not necessarily an integer number of bits per symbol).
  • the B bits input to modulus encoder 204 are processed such that they are represented by an integer R:
  • R b 0 + b,2 + b 2 2 2 + . . . + b B -.2 B - 1 , where b 0 is the first bit in time and b B .., is the last bit in time of the B bits submitted to modulus encoder 204.
  • b 0 is the first bit in time
  • b B .. is the last bit in time of the B bits submitted to modulus encoder 204.
  • values are either zero or one in this expression.
  • R may be suitably expressed in terms of a hexadecimal number for processing by a microprocessor resident at encoder 200. The number R is further processed such that it is represented in terms of the reordered moduli (M 0 '-M 5 '):
  • modulus encoder 204 The output of modulus encoder 204 are the six index values designated by K 0 '-K 5 ', and 0 ⁇ K-' ⁇ M,' . Modulus encoder 204 performs the following operations to determine these values:
  • K 0 ' RO modulo M 0 ';
  • R1 (R0-K 0 ')/M 0 ';
  • , ' R1 modulo MJ;
  • R2 (RI -KJJ/M ;
  • K 2 ' R2 modulo M 2 ';
  • R3 (R2-K 2 ')/M 2 ';
  • modulus encoder 204 may perform a conventional MMC procedure using the B bits and M 0 '-M 5 ' as respective inputs. As shown in FIG. 2, the i' values are utilized for inputs to an index reordering element 221 . Index reordering element 221 is operationally associated with moduli reordering element 220 such that the appropriate indices are utilized with the corresponding signal point maps.
  • index reordering element 221 functions to reorder the K j ' indices in a manner corresponding to the function performed by moduli reordering element 220. For example, if moduli reordering element 220 performs the following reordering function:
  • index reordering element 221 preferably performs the following reordering function:
  • Index reordering element 221 may also be employed where multiple moduli are reordered, e.g., for the above example where RBS occurs at symbol phases 1 and 3, index reordering element 221 may reorder the indices according to the following scheme:
  • the reordered K, values are used for inputs to the respective signal point maps 208, 21 0, 21 2, 214, 21 6, and 21 8.
  • a K, value serves as an index for the selection of a particular codeword from the corresponding M, map.
  • the respective outputs from signal point maps 208, 21 0, 21 2, 214, 21 6, and 21 8 are universal PCM codewords that are related to the ⁇ -law (or A-law) signal point constellations employed by telephone network 1 30.
  • the specific values assigned to these universal codewords may be found in the proposed V.PCM Standard, published December 1 2, 1 997 by the ITU Telecommunications Standardization Sector. The entire content of this publication is incorporated herein by reference.
  • these universal PCM codewords (identified by U, in FIG. 2) are transmitted in a temporal order corresponding to their subscript designations.
  • the known scheme causes U 0 , which is represented by the PCM 0 octet, to be transmitted first in time and U 5 , which is represented by the PCM 5 octet, to be transmitted last in time.
  • the universal codewords U may be routed to a polarity block 224, which preferably functions to assign a positive or negative sign bit to the universal codewords.
  • the particular sign assigned to the individual universal codewords may be dictated by a spectral shaper 228 that communicates with polarity block 224.
  • Spectral shaper 228 may use redundant sign bits to control the spectral content of
  • Polarity block 224 produces a PCM codeword associated with each universal codeword.
  • PCM 0 refers to the first-transmitted symbol or octet in the data frame and
  • PCM 5 refers to the last-transmitted symbol or octet in the data frame.
  • corresponds to U, for all values of /.
  • the PCM octets may be input to a multiplexer 232, which is suitably controlled to produce a serial stream of the PCM octets in the designated temporal order.
  • the PCM octets are transmitted through telephone network 1 30, converted into analog signals at, e.g., line card 140, and received by user modem 1 60 (see FIG. 1).
  • User modem 1 60 may employ suitable analog or digital filtering, conditioning, decoding, and processing techniques to obtain the original data (indicated by b 0 - b B .. in FIG. 2) .
  • user modem 1 60 preferably includes a moduli reorder element and an index reorder element that perform the reverse operation of moduli reordering element 220 and index reordering element 221 , respectively.
  • the particular reordering operations are preferably predetermined by the modem system during the initialization procedure and stored at ISP modem 105 and user modem 1 60.
  • the signal points within any given constellation are suitably arranged to address certain transmit power considerations.
  • a currently proposed 56 kbps protocol orders the universal codewords in descending order relative to the corresponding index value.
  • One exemplary prior art constellation arrangement is set forth in Table 1 below.
  • lower universal codeword values correspond to PCM codewords having lower signal magnitudes and higher universal codeword values correspond to PCM codewords having higher signal magnitudes.
  • Table 1 in the prior art arrangement, the lowest value of K identifies the universal codeword corresponding to the largest signal point magnitude.
  • the lower index values for any given constellation may have a higher likelihood of occurrence than the higher index values. This occurs because 2 B ⁇ M 0 M, . . .M L ., , and because some of the larger index values may not be selected as often in response to the B-bit input to modulus encoder 204.
  • One preferred embodiment of the present invention suitably reorders the universal codewords in an ascending order (see Table 1 ). This index ordering assures
  • transmit power is conserved and, preferably, optimized by the particular constellation ordering scheme.
  • the present invention may suitably arrange the order of the universal codewords in response to an adaptive analysis performed during an initialization period.
  • Such an adaptive analysis may contemplate any number of cost functions in addition to transmit power.
  • Those skilled in the art will be familiar with various operating characteristics and functional limitations that may require optimization and which may be affected by the particular constellation ordering.
  • transmit power considerations can play an important role during adaptive signal point constellation design and various optimization techniques.
  • Current FCC regulations limit the total average transmit power over a telephone network channel to -12dBm.
  • modem systems must be designed to operate in a robust and reliable manner while maintaining the total average transmit power within the FCC limit. For this reason, prior art modem systems may estimate the total average power associated with a given signal point constellation and compare the estimated value with the -1 2dBm limit. If the estimated power exceeds the FCC limit, then the modem system may redesign the constellation such that its total average power is within an acceptable range.
  • the present invention also contemplates the transmission of symbols having an effective non-integer bit count.
  • 2 B ⁇ M 0 M, . . . M L . , .
  • the transmission of fractional bits per symbol makes an exact power determination more difficult, especially where MMC techniques are employed. This difficulty is due to the unequal probability of occurrence associated with the signal points within the signal point constellations.
  • the present invention provides a methodology for determining the exact average transmit power for the modem system described herein. Generally, the power calculation initially determines the probability of transmission of each individual signal point within a particular constellation. Then, the probabilities of occurrence are used to generate the average total power for each of the constellations (six in the exemplary embodiment described herein).
  • the power for all of the constellations is determined and averaged to produce the total average power for the given data communication session.
  • the counting scheme presented herein is described herein for explanatory purposes and to conveniently set forth the principles of the present invention. It should be appreciated that variations of the total average power derivation are also contemplated by the present invention.
  • modulus encoder 204 receives an input of B bits (B may vary depending on the current data rate).
  • B bits can be represented as an integer
  • the theoretical total transmission power can be determined by summing the power of the six PCM levels mapped by modulus encoder 204 from each possible B-bit pattern, then dividing the sum by 2 B .
  • the total average transmission power may be obtained by first summing, for all integer values of R, the energy of the six symbols indexed by modulus encoder 204. The total transmission power is then obtained by dividing this sum by 2 B *6, because there are a total of 2 B terms and each term includes six symbols.
  • the procedure described immediately above may be too computationally rigorous to implement in an efficient and practical manner.
  • the present inventors have recognized that the MMC procedure lends itself to an elegant "shortcut" technique for determining the exact transmission power associated with given MMC mapping parameters.
  • This simplified approach takes advantage of the observation that the MMC procedure represents the integer R as a "number" between 000000 to AoA- ⁇ A g A ⁇ s, using M 0 -M 5 as the mixed counting bases, where A 0 A .A 2 A 3 A 4 A 5 is the output of modulus encoder 204 corresponding to an input pattern of all B bits equal to one.
  • the counting scheme may be expressed as shown in Table 2 below.
  • each column (representing the values of the K, indices) includes 2 B entries or index values.
  • the index number begins with 0, counts up to M,-1 , returns to 0, and continues counting in this manner until it reaches the final count of A,.
  • the index value K 0 changes once for every increment of R.
  • the index value K 2 changes once for every M 0 M, increments of R.
  • the counting scheme follows this general trend for all of the bases and moduli.
  • the frequency of occurrence for each number in a particular column corresponds to the probability of occurrence of the particular signal point identified by that index number.
  • the total number of occurrences (identified by columns in Table 2) for any K is equal to 2 B .
  • the probability of occurrence for the various indices can be uniquely defined by the values of M, and 2 B or uniquely defined by the MMC output of the integer 2 B -1 , i.e., a B-bit pattern of all ones.
  • the respective values of A may be equal to any integer from 0 to M,-1 . Accordingly, depending upon the specific value of Aava higher values of K, may not occur as often as relatively lower values of K,. This characteristic is illustrated in Table 2, which shows values of A, occurring in the final "count cycle" from 0 to M,-1 .
  • the value of A, relative to M,-1 affects the probability of occurrence associated with the different K, index values (described in more detail below).
  • the present invention defines and determines several parameters to facilitate the determination of the exact transmit power for a given constellation set.
  • the parameter P k may be defined as: k
  • P k represents a “full counting cycle” within the r-th position or column.
  • a “full counting cycle” means
  • 1 8 one counting iteration from 0 back to 0 for the /-th modulus.
  • the index value of K 0 repeats after M 0 counts (the count may be 0, 1 , 2, . . . M 0 -1 , 0, 1 , 2, . . . M 0 -1 , 0 . . .).
  • the index value of K repeats only after M Q M - I counts.
  • the parameter q k provides the number of "full” counting cycles associated with the k-t position as the MMC output counts up from all zeros to the highest value defined by the A k values.
  • the INT operation ensures that only full counting cycles are accounted for.
  • This parameter provides an indication, for the k-th position, of the number of "sub-cycles" contained in the residual number of index values remaining after considering the "full" counting cycles defined by q k .
  • a "sub-cycle" for the k-th position equals P k .
  • a full cycle equals M 0 M
  • counts and a sub- cycle equals M 0 counts. It should be appreciated that the values of r k will be dependent upon the respective values of A k .
  • the ending count associated with a particular A k value may correspond to an integer number of full counting cycles, i.e., P k cycles, or an integer number of full sub-cycles, i.e., P k ⁇ cycles, or a residual index count associated with a sub-cycle.
  • the MODULO operation provides this residual amount, which may be considered to be the remainder associated with r k .
  • FIG. 3 depicts an exact power determination process 300 that may be performed by the modem system during an initialization procedure. Process 300 may be repeated for any number of signal point constellation sets and process 300 may
  • process 300 may implement process 300 in accordance with any number of conventional techniques and that process 300 may be carried out by any number of microprocessor elements, memory elements, and the like, located at ISP modem 105 and/or user modem 1 60.
  • Process 300 may begin with a task 302, which obtains the current values for B (number of bits encoded for each frame) and all M
  • a task 304 is preferably performed to determine a number of parameters and variables used by process 300 in its computation of the exact total average transmit power. In particular, task 304 determines P k , q k , r k , and c k for all values of k.
  • k indicates the k-t module or constellation and / indicates a specific index value associated with the k-th constellation. Accordingly, the maximum value of / is equal to the number of signal points contained in the associated constellation.
  • query task 310 leads to a query task 314.
  • a task 320 may be performed to suitably save the value.
  • a query task 322 may be performed to check whether the current value of / is the maximum for the given constellation. If not, then a task 324 may increment / and cause process 300 to repeat the probability calculation for the incremented value of / ' . If, however, query task 322 determines that the maximum value of / has been reached, then a query task 326 may be performed to test whether the current value of k is the maximum for the given set of constellations. If not, then a task 328 may increment k and cause process 300 to repeat the probability calculations for all of the signal points associated with the next constellation. If the maximum value of k has been reached, then query task 326 may lead to a task 330. At this time, process 300 has collected and stored the probabilistic data associated with the transmission occurrence of the various signal points contained in each of the designated constellations.
  • Task 330 suitably determines the average power for the k-th module (for all k). In the preferred embodiment, task 330 performs the following operation: ⁇ f.-l
  • E k is the average power for the k-th module, i.e., constellation
  • S k l is the magnitude of the i-th signal point in the k-th module.
  • L is the number of symbols per data frame, e.g., six in the example presented herein.
  • the different probability expressions may be included in a generalized power determination that may be derived without directly obtaining the individual probability values.
  • the determination of the three probability values may be contemplated by such a generalized power expression yet be "invisible" from a computational perspective.
  • the exact power determination can be a rigorous and complex procedure. Accordingly, it may be desirable to implement alternative techniques that provide an effective estimate of the total average power in a computationally efficient manner. For example, in applications using the ascending signal point numbering scheme described above in connection with Table 1 , the lower magnitude signal points may occur more often than the higher magnitude signal points. Therefore, a total average power estimate using a uniform probability for all signal points serves as an upper bound. For example, one preferred embodiment may be configured to perform the following power estimate rather than the foregoing exact power calculation:
  • a calculation based on an assumption of uniform probability may not result in a valid upper bound.
  • an assumption of uniform probability may result in a total power estimate that is lower than the actual total power. This may occur because the actual probability of transmission of the higher magnitude signal points may be higher than any uniform probability estimate.
  • the present invention may also be configured to perform an estimated power determination process 400 to accommodate such descending signal point numbering arrangements.
  • Process 400 begins with a task 402, which obtains the parameters B, M 0 -M 5 , and the signal point magnitudes associated with the constellations.
  • Task 402 is similar to task 302 described above in connection with FIG. 3.
  • a task 404 may be performed to determine the value of P k for all k values.
  • Task 404 is similar to task 304, and the expression for P k is set forth above in connection with the description of task 304.
  • a task 406 is preferably performed to calculate the probabilities of transmission associated with the given signal points.
  • the upper bound may be determined by assuming that the probability of transmission for all signal points will be the highest of the three probability expressions set forth in tasks 312, 31 6, and 318 and described in detail above.
  • process 400 utilizes the probability expression set forth in task 31 2 to generate the probabilities for all signal points. Again, this expression is:
  • Task 408 is then performed to estimate the total average transmit power using the above probability expression for all values of / ' .
  • Task 408 performs the following calculation:
  • the above upper bound should be a tighter upper bound than t -i
  • the present invention provides improved techniques for use in a modem system that employs a multiple modulus conversion (MMC) procedure.
  • MMC multiple modulus conversion
  • the signal point constellations are arranged to receive the output indices generated by the MMC procedure such that the probability of transmission of higher power codewords is relatively low compared to the probability of transmission of lower power codewords, while maintaining the
  • the present invention provides techniques for obtaining the probability of occurrence for each constellation or signal point associated with a given data communication session and for determining the theoretical average transmit power associated with a given set of signal point constellations and MMC parameters.
  • the present invention provides a method for estimating a reasonable upper bound for the average transmit power associated with a given set of signal point constellations and MMC parameters.
  • the present invention is also capable of transmitting the codewords defined by an MMC procedure in a specific order that may be optimized in accordance with a particular cost function associated with the modem system.
  • process 300 need not calculate a probability value for each increment of / ' ; it may simply calculate the three values and apply the appropriate value depending upon which condition the current value of / satisfies. Such changes and modifications are intended to be included within the scope of the present invention, as expressed in the following claims.

Landscapes

  • Physics & Mathematics (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Digital Transmission Methods That Use Modulated Carrier Waves (AREA)

Abstract

Un système de modem MIC 56 kbps utilise des techniques de conversion de modules multiples (MMC) pendant le codage et le décodage. Les techniques MMC peuvent être utilisées pour le codage d'un certain nombre de bits d'entrée sous forme d'indices de point de signal associés à plusieurs constellations de points de signal, le nombre de points de signal dans une constellation donnée étant relatif au module utilisé pendant le codage MMC. Le système de modem comprend un élément de réordonnancement de module, qui réordonnance les modules en fonction de la manière dont le dispositif de cartographie MMC fonctionne. L'élément de réordonnancement peut être configuré de sorte qu'il traite les problèmes de puissance d'émission et de dégradation numérique. Outre le réordonnancement de module, les points de signal individuels dans une constellation donnée peuvent être réordonnancés, de sorte que les considérations de puissance d'émission soient traitées. Ledit réordonnancement des points de signal peut permettre que les points de signal ayant une puissance d'émission supérieure apparaissent moins fréquemment dans une session de communication pratique. La technique MMC peut également étendue pour le calcul ou l'estimation précise de la puissance d'émission moyenne totale associée à une session de communication.
PCT/US1999/001559 1998-01-26 1999-01-26 Techniques de conversion de modules multiples pour modem mic WO1999038299A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US09/013,671 1998-01-26
US09/013,671 US6034991A (en) 1998-01-26 1998-01-26 Method and apparatus for implementing enhanced multiple modulus conversion techniques in a signal point mapping context

Publications (1)

Publication Number Publication Date
WO1999038299A1 true WO1999038299A1 (fr) 1999-07-29

Family

ID=21761118

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US1999/001559 WO1999038299A1 (fr) 1998-01-26 1999-01-26 Techniques de conversion de modules multiples pour modem mic

Country Status (2)

Country Link
US (2) US6034991A (fr)
WO (1) WO1999038299A1 (fr)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2001013596A1 (fr) * 1999-07-31 2001-02-22 Altocom, Inc. Correction numerique d'incidents pour modem v.90
WO2001019045A1 (fr) * 1999-09-10 2001-03-15 Conexant Systems, Inc. Modem pcm avec precodage et pre-egalisation
US7190727B1 (en) 1999-07-31 2007-03-13 Broadcom Corporation Categorization of impairments affecting a communication channel
US7203248B1 (en) 1999-07-31 2007-04-10 Broadcom Corporation Impairment sensitive selection of constellation points for communication across a channel

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6381266B1 (en) * 1998-09-30 2002-04-30 Conexant Systems, Inc. Method and apparatus for identifying the encoding type of a central office codec
FR2784528A1 (fr) * 1998-10-13 2000-04-14 Koninkl Philips Electronics Nv Methode de construction d'un ensemble de constellations destine a etre utilise pour transmettre des donnees entre un emetteur et un recepteur
US6553063B1 (en) * 1998-10-30 2003-04-22 Broadcom Corporation Constellation-multiplexed transmitter and receiver
US7254167B2 (en) 1998-10-30 2007-08-07 Broadcom Corporation Constellation-multiplexed transmitter and receiver
US6711207B1 (en) * 1999-03-11 2004-03-23 Globespanvirata, Inc. System and method for reduced power consumption in digital subscriber line modems
US7339996B2 (en) * 1999-06-24 2008-03-04 Intel Corporation Receiver codec super set constellation generator
US6721363B1 (en) * 1999-06-24 2004-04-13 Intel Corporation Receiver CODEC super set constellation generator
US7366470B1 (en) 1999-06-24 2008-04-29 Intel Corporation Inter-modulation distortion compensation
US6724814B1 (en) 1999-06-24 2004-04-20 Intel Corporation Pad and CODEC detection
US6721357B1 (en) * 1999-06-24 2004-04-13 Intel Corporation Constellation generation and re-evaluation
US7072387B1 (en) * 1999-12-15 2006-07-04 Paradyne Corporation Fractional bit rate encoding in a discrete multi-tone communication system
US6266376B1 (en) * 2000-02-04 2001-07-24 Motorola, Inc. System and method for adjusting PCM data frames for robbed bit signaling in a telephone network
US6434190B1 (en) 2000-02-10 2002-08-13 Texas Instruments Incorporated Generalized precoder for the upstream voiceband modem channel
US7180916B2 (en) * 2001-04-27 2007-02-20 Agere Systems Inc. Compensating for random robbed bit signaling
US6697831B2 (en) * 2002-02-28 2004-02-24 Ess Technology, Inc. Device and method for performing multiple modulus conversion using inverse modulus multiplication
US7111226B1 (en) * 2002-05-31 2006-09-19 Broadcom Corporation Communication decoder employing single trellis to support multiple code rates and/or multiple modulations
JP3947876B2 (ja) * 2004-02-03 2007-07-25 Ｎｅｃアクセステクニカ株式会社 Ｐｃｍ符号を用いたデータ伝送システムおよび方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5103227A (en) * 1990-09-26 1992-04-07 At&T Bell Laboratories Modulus converter for fractional rate encoding
WO1992017972A1 (fr) * 1991-03-28 1992-10-15 British Telecommunications Public Limited Company Synchronisation de donnees
EP0725487A2 (fr) * 1995-02-01 1996-08-07 AT&T Corp. Codage à niveaux multiples pour bits fractionnairs
US5598435A (en) * 1993-12-23 1997-01-28 British Telecommunications Public Limited Company Digital modulation using QAM with multiple signal point constellations not equal to a power of two
EP0852438A2 (fr) * 1996-12-04 1998-07-08 Rockwell Science Center, Inc. Mise en forme de treillis pour transmission à plusieurs niveaux, lesdits niveaux étant accordés aux niveaux de quantisation d'un réseau
WO1998039883A1 (fr) * 1997-03-03 1998-09-11 3Com Corporation Procede de signalisation utilisant la conversion par modules multiples et le mappage des coquilles

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5056112A (en) * 1989-07-28 1991-10-08 At&T Bell Laboratories Interleaving in coded modulation for mobile radio
US5230010A (en) * 1990-09-26 1993-07-20 American Telephone & Telegraph Company Fractional rate modulation
US5222077A (en) * 1991-04-09 1993-06-22 Racal-Datacom, Inc. Radix mapping with variable number of symbols in mapping period
CA2107452C (fr) * 1992-10-30 1998-09-29 William Lewis Betts Systeme d'optimisation pour canaux de communication
US5684834A (en) * 1993-06-14 1997-11-04 Paradyne Corporation Simultaneous analog and digital communication using fractional rate encoding
US5428641A (en) * 1993-07-23 1995-06-27 Motorola, Inc. Device and method for utilizing zero-padding constellation switching with frame mapping
GB2309363B (en) * 1996-01-17 2000-07-12 Motorola Ltd Multicarrier communication system and method for peak power control
US5825816A (en) * 1997-02-14 1998-10-20 General Datacomm, Inc. Spectral and power shaping mapper for high data rate signalling
US5838724A (en) * 1997-02-14 1998-11-17 General Datacomm, Inc. Spectral and power shaping mapper for high data rate signalling
US5862184A (en) * 1997-02-14 1999-01-19 General Datacomm, Inc. Mapper for high data rate transmission through channels subject to robbed bit signalling
US6084915A (en) * 1997-03-03 2000-07-04 3Com Corporation Signaling method having mixed-base shell map indices
US6084883A (en) * 1997-07-07 2000-07-04 3Com Corporation Efficient data transmission over digital telephone networks using multiple modulus conversion
US6065030A (en) * 1998-03-30 2000-05-16 Conexant Systems, Inc. Method and apparatus for implementing short-word division techniques in a multiple modulus conversion context

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5103227A (en) * 1990-09-26 1992-04-07 At&T Bell Laboratories Modulus converter for fractional rate encoding
WO1992017972A1 (fr) * 1991-03-28 1992-10-15 British Telecommunications Public Limited Company Synchronisation de donnees
US5598435A (en) * 1993-12-23 1997-01-28 British Telecommunications Public Limited Company Digital modulation using QAM with multiple signal point constellations not equal to a power of two
EP0725487A2 (fr) * 1995-02-01 1996-08-07 AT&T Corp. Codage à niveaux multiples pour bits fractionnairs
EP0852438A2 (fr) * 1996-12-04 1998-07-08 Rockwell Science Center, Inc. Mise en forme de treillis pour transmission à plusieurs niveaux, lesdits niveaux étant accordés aux niveaux de quantisation d'un réseau
WO1998039883A1 (fr) * 1997-03-03 1998-09-11 3Com Corporation Procede de signalisation utilisant la conversion par modules multiples et le mappage des coquilles

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
HODGKISS B: "DSP IMPLEMENTATION CONSIDERATIONS IN HIGH-SPEED TELEPHONE MODEMS", DSP. CONFERENCE PROCEEDINGS, 3 December 1997 (1997-12-03), pages 77 - 84, XP000199773 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2001013596A1 (fr) * 1999-07-31 2001-02-22 Altocom, Inc. Correction numerique d'incidents pour modem v.90
US7190727B1 (en) 1999-07-31 2007-03-13 Broadcom Corporation Categorization of impairments affecting a communication channel
US7203248B1 (en) 1999-07-31 2007-04-10 Broadcom Corporation Impairment sensitive selection of constellation points for communication across a channel
US7397845B1 (en) 1999-07-31 2008-07-08 Broadcom Corporation Impairment compensation sequence for communication
WO2001019045A1 (fr) * 1999-09-10 2001-03-15 Conexant Systems, Inc. Modem pcm avec precodage et pre-egalisation
US6414989B1 (en) 1999-09-10 2002-07-02 Conexant Systems, Inc. Upstream PCM transmission for a modem system

Also Published As

Publication number Publication date
US6438158B1 (en) 2002-08-20
US6034991A (en) 2000-03-07

Similar Documents

Publication Publication Date Title
US6034991A (en) Method and apparatus for implementing enhanced multiple modulus conversion techniques in a signal point mapping context
EP1076987B1 (fr) Procedes et appareil permettant de verifier des niveaux de puissance d'emission dans un systeme de transmission limite a point de signal
US6084915A (en) Signaling method having mixed-base shell map indices
CA2167746C (fr) Codage multiniveau pour bits fractionnaires
US5838724A (en) Spectral and power shaping mapper for high data rate signalling
US6084883A (en) Efficient data transmission over digital telephone networks using multiple modulus conversion
US5825816A (en) Spectral and power shaping mapper for high data rate signalling
US5280503A (en) Data communication system with data rate throttling
US6081555A (en) Methods and apparatus for implementing shell mapping techniques in the context of a PCM-based modem communications system
EP1038378A1 (fr) Dispositif et procede de precodage de signaux de donnees destines a la transmission mic
EP0701352A2 (fr) Adaptation automatique de prédistorsion pour la transmission sur canaux non-linéaires
EP0533363A2 (fr) Codeur et décodeur non-linéaire pour transmission d'information par canaux non-linéaires
US5818879A (en) Device, system and method for spectrally shaping transmitted data signals
US6065030A (en) Method and apparatus for implementing short-word division techniques in a multiple modulus conversion context
JP3527508B2 (ja) フレーム・マッピングを用いてゼロ・パディング・コンステレーション切り替えを利用するための装置および方法
WO1999034566A1 (fr) Systeme, dispositif et procede de transmission en amont mic utilisant une constellation de transmission optimisee
EP0880838A1 (fr) Technique de communication utilisant un code en treillis etabli a partir d'un ensemble fini de points de signal bande de base
US7251270B2 (en) Systems and methods for fractional bit rate encoding in a communication system
US6993067B1 (en) Fractional bit rate encoding in a pulse amplitude modulation communication system
CA2267362A1 (fr) Dispositif, systeme et procede pour former le spectre des signaux de donnees emis
US5995548A (en) Signaling method using multiple modulus shell mapping
US6553074B1 (en) Method and device for combating PCM line impairments
WO1998039883A1 (fr) Procede de signalisation utilisant la conversion par modules multiples et le mappage des coquilles
EP1031217B1 (fr) Procede et dispositif d'optimisation des techniques de mise en correspondance d'enveloppe par utilisation d'une fonction approchee cout/puissance
US7339996B2 (en) Receiver codec super set constellation generator

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): CA JP NO

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE

121 Ep: the epo has been informed by wipo that ep was designated in this application
DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
122 Ep: pct application non-entry in european phase